The informally built space as a

means to achieve life chances

Ana Cludia Cardoso
Departamento de Arquitetura e Urbanismo, Universidade Federal do Par, Conj. Euclides
Figueiredo Rua B, no. 6 cod 66620-640, Belm PA / Brazil, phone 55 91 81120604, e-mail:
cardoso@ufpa.br

Urban form studies and squatter settlements

Brazilian official institutions, following the frequent pattern of developing countries urban
mismanagement, have not succeeded in wrestling with Brazils historic tradition of occupation
prior to tenure regularization and of gradual (non-planned) expansion of urban areas (Marx,
1991). This tradition favoured the production of informal settlements, which are increasingly
shaping cities expansion. Little is known about how much the informal origin of consolidated low
income areas interferes with their integration into the whole city, and about the contribution of
formerly informal space to its inhabitants life chances over time, as the transformations of
consolidation take place.

This process is particularly easy to be observed in Belm, Brazil, the city taken as case study in
this paper. Its historic background calls attention to how insufficient were traditional policies to
tackle demands for land, housing and infrastructure provision. The city is 388 years old, was
funded in a site surrounded by flood plains and used to be the biggest city and the main port in
the Brazilian Amazon. Social inequalities were established since colonial times, and reinforced
during the economic boom of rubber exploitation (1850-1920) and since the launching of official
programmes to populate and develop the region (1960s). Along the last fifty years the city 's
flood plains were occupied by hybrid typologies of informal settlements, a combination of
squatter settlements, clandestine settlements and family-by-family occupations1. Over a period,
they were usually upgraded, receiving social (education and health care facilities) and physical
infrastructure (water supply, sanitation facilities, drainage, urban roads and solid disposal
facilities), and transformed into low-income districts, completing a cycle of consolidation (Gilbert
& Guggler, 2000). Nevertheless, this process was uneven, mostly done though inhabitants
effort, and the rhythm, quality, and quantity of infrastructure provided, clearly affected the
enhancement of their life chances over the corresponding period.

The informal process of low income housing production is very different from official social
housing production. The former occurs in a fragmented fashion, through the cumulative action of
several agents, ranging from informal developers, to inhabitants, to the government. It usually
relies on a quite flexible relationship with time, without a hurry to meet the requirements of urban
infrastructure and service provision. It does not compromise with formal technical and legal

1
A discussion about informal settlements typologies is presented in Burgess (1985).
standards or previous quantification of land use and density of occupation. While the latter,
searches for affordable locations, needs to meet conventional infrastructure requirements at the
outset, and is programmed according to previously established standards of occupation and
technical paradigms.

Studies of the physical form of informal settlements are scarce, and usually focused on plot
dimension in order to investigate tenure ownership and density relationships with physical form.
For instance, Paynes (1977) case study carried out in Delhi had its physical analysis
concentrated on the plot, in order to identify typologies related to tenure ownership, amount of
space per land use, and efficiency ratios (between settlements public space and space
appropriated by the community and number of units). Other examples are provided by Correas
(1985), Doshis (Curtis, 1988; Steele, 1998) and the Centre for Minimum Cost Housing of McGill
University investigations. Correa discussed plot arrangements and building typologies aiming to
achieve different scales of common space (from small courtyards to the street), and densities
responsive to the climatic and socio-economic Indian environments, and were applied in
Belapur, New Bombay. Doshi carried out research on informal settlement space through the
Vastu-Shipa Foundation, during the 1980s, and applied the findings to the Aranya housing
project in Ahmedabad (Bhatt & Scriver, 1990). The series How the Other Half Builds, published
by the cited McGill University centre, investigated characteristics of plots and public spaces in
informal settlements and searched for new design methods for the production of responsible
housing for the poor (Bhatt, 1990).

The Indian experience is a particular example of investigation of informal settlement space as a

means to enhance official settlements spatial qualities. But evidence of gentrification in the new
building schemes diminished the stimulus to further research (Bhatt & Scriver, 1990). The focus
point was the production of housing through new settlements, although Payne (1977) and
Correa (1985) recognised that the problem in the developing countries is not of insufficient
housing (the poor always succeed in producing housing), but of precarious integration of
informal settlements into the existing city. This is the point from where this article starts.

In Belm, the spatial strategies of the poor are conspicuous, regarding to the compromise of
location's and street pattern's choices with aims such as: a) control of housing location, type and
cost, b) conditions to minimise transport expenses and favour to access to income; c)
favourable prospects of infrastructure provision (Payne, 1977: 194). Poor inhabitants need of
preserving identity, independence and decision-making freedom, and of struggling for legal
political rights is also visible (Wratten, 1995:15). Within its informal settlements, action of
agents other than inhabitant occurs at the late stages of consolidation and is dependent on
possible political and economic advantages to the external agent; such circumstances has
allowed the tracking of how much inhabitants are able to do by themselves and highlighted the
insufficiency of traditional top-down approaches of official land, housing and infrastructure
provision, and the lack of knowledge to explain the everyday process of informal settlements
production and to correctly interfere in their cycle of assimilation by the pre-existing city
(Cardoso, 2002a).

According to Moudon (1997) it is possible to assess patterns of density and land use over time,
and to understand the socio-economic conditions of a settlements origins, by analysing the
combination of basic components of urban form, according to different resolutions and over a
selected period of time. Physical elements evolve diversely from city to city and even within a
city, according to the socio-economic and cultural profile of successive users. This evolution

2
creates cycles of transformation, which are able to show rates of functional (related to activities)
and physical change, correspondent to the citys economic and cultural conditions.

Most typologies of low-income informal settlements are constituted by the same fundamental
physical elements of urban form as any formal space: buildings and their related open spaces,
plots, and streets. These settlements also have an urban form, which can be observed on
several scales of interface (such as building/plot, street/block, district/city), and are in continuous
transformation. They thus enable the basic principles of all morphological approaches to
understanding a city or town through its physical urban form to be used (Moudon,1997:7).

It was not possible to fully adopt the traditional research method used by morphology schools, to
pursue the aim of carrying out an empirical research compromised with the relationship between
spatial transformation of informal settlements and their inhabitants' enhancement of life
chances, because, as usual, there were no series of data available to allow study of their
development over time; the urban form to be analysed here has either been recently generated
or is in a clear process of transformation, shaped according to the socio-economic conditions of
spaces most active producer, the land invasion inhabitant. To overcome this methodological
difficulty, different areas were taken as part of a case study in order to compare their different
stages of evolution and to extract findings, possibly elucidative of them all. The case study
areas selection was based on their similar physical and socio- economic contexts, their capacity
to represent different locations within the city, and different stages of consolidation through their
configuration (here taken as a snapshot of a moment in a settlements development) and time
scale.

Searching for a spatial dimension of life chances

The approach introduced here is based on the premises of urban morphology and on the
concepts of life chances and contemporary poverty. Cardoso (2002a, 2002b) has presented a
discussion linking these two concepts, which now is extended to the production of informal
settlements built space and to the potentials to enhance inhabitants long term prospects. From
Dahrendorf (1977, 1988), life chances is defined as the long-term prospects of someone
brought about by choices from available options and according to a persons social objectives,
and subdivided in three constitutive elements a) entitlements (legal means of access), b)
provision (material goods and services) and c) ligatures (values and motivations that make
sense to choices made by someone). Kempen (1994) defines poverty as lack of life chances;
the modern poor are mainly those who are prevented from accessing, among other basic
resources, education and health care, remunerated work, housing and a safe environment, and
have been split from their social context (Moser, 1998). The concept of life chances offers a
broad structure to approach social issues, which was found compatible with the recent
discussion on poverty (to be assessed in qualitative and quantitative terms) and also with the
unregulated context of agents action within informal settlements; the association of these two
concepts allowed a multidimensional perspective to the investigation on informal settlements
production and evolution (based on spatial and social variables) (Cardoso, 2002a).

This theoretical framework assumes the importance of inhabitants point of view about their
circumstances and the understanding of their motivations (ligadures) to settle somewhere,
despite provision conditions (of housing, land, infrastructure); furthermore, the addition of how
these choice are expressed in physical access terms (entitlements) adds light to the recognised
but not so well understood cycle of consolidation, or process of informal settlements
amalgamation into the pre-existing city (Gilbert &Guggler, 2000).

3
In order to be consistent with the assumptions presented, this approach seeks for a method to
accurately describe and analyse settlements pattern of formation and development over time,
through intensification of occupation. Informal settlements have often been considered chaotic
in relation to conventional architectural paradigms of order and harmony, indicating the need for
more appropriate approaches to investigate these settlements spatial form. Urban
morphological approaches appropriateness to investigate underlying structures as much as
visible forms of urban growth and change was the reason for their usage, to asses a spatial
dimension of life chances, through their tools of analysis, supplemented by elements of the
planning discipline, space syntax techniques and by investigation of a sites natural attributes.
Overall, the method presented hopes to contribute to the investigation and management of the
consolidation process in cities where it occurs.

Up to this stage of research development, the spatial assessment of entitlement, provision and
ligatures is mostly based on the relationship of streets with the other elements of the urban form,
because they are the most stable and prevalent element in these areas. Over cycles of urban
change, streets are the first spatial elements of urban occupation, providing access to plots and
defining blocks through their subdivisions. They are intermediaries between private and
collective scales and aggregate investments in ownership and infrastructure that make them
more resistant to change than plots and buildings. In formal contexts, plots have a legal
definition in maps, which safeguard their longevity; however, they are objects of subdivision and
amalgamation according to market conditions and interests. Buildings, in their turn, might be
replaced or refurbished many times over the life-cycle of a plot, since they depend basically on
the level of control the owner has over his/her plot, and on the amount of resources available
(Cannigia & Maffei, 1995; Brand, 1997; McGlynn, tutorial 2002).

Despite the risk of isolating the different dimensions of investigation from each other too much, it
is assumed that spatial entitlements may be assessed through the potential to displacement
within a city offered by the urban fabric, particularly the street layout. Spatial provision is
constituted by the arrangement of streets, blocks, plots and buildings and by infrastructure. And
spatial ligatures are observed through watching the patterns of use of public spaces. The
particular approach to each of these elements was chosen according to its capacity to express
spaces complexity by encapsulating, for instance, considerations of global and local scales,
public and private dimensions, and of space and society. It is important to recognize the
continuum between these extremes to minimize inappropriate clinging to outdated static
positions, and the risk of losing the dynamic of observed process (Arida, 1998).

Space and entitlements

Architectural research has demonstrated that space is neither the determinant of, nor neutral to,
society; instead it shapes possibilities (Hillier, 1996:206). Then, if entitlement is defined as a
socially acceptable means of access to desired goals, it must have a spatial dimension. In
theory, international agreements about human rights and national constitutions are recognised
by local laws and respected as a legal guarantee of inhabitants basic rights. However, within
the reality of a developing country city, the uneven process of urbanisation differentiates space
and access conditions to, for instance, housing, schools and health care, in quantitative
(availability) and qualitative (reliability) terms.

Socio-economic constraints on the realisation of basic human rights (e.g.: rights of access to
shelter, education, income and health care) are evident within the developing country city to the

4
extent to which its low-income inhabitants are dependent on informal housing and the informal
economy (Hall, 1987:251). Physical access to and availability of land are crucial for those who
decide to settle in the city for whatever reason; the absence of policies to deliver land prior to its
occupation, and the high costs of the formal market, determine low income inhabitants choice
(Payne, 1999). The poorer this population, the more it needs good accessibility, preferably
location near to the city centre. In order to accomplish that, the poor search for flooded or
prone-to-slide land, the unsuitable land which has been avoided by the formal market, in the
adjacencies of urban centres, in which to settle themselves (Hardoy & Satterthwaite, 1987:
306).

Despite many technological changes in the Western citys way of life, the informal settlements
inhabitants still seek for basic urban services and facilities, and need to benefit from short
distances, highly used spaces and superimposition of activities. Ironically, these inhabitants low
incomes push them down to unsuitable sites, which, when recently occupied, usually are among
the most segregated and deepest spaces within the city. However, the weak connection of
informal settlements to the formal city, or to its sub centres, may be just temporary, depending
on restrictions imposed by the sites physical conditions, which may be transformable over time,
and the capacity of a new grid to be embedded in the existing one (Hillier et al., 2000). This
allows the creation of more connections and a better potential for entitlements and justifies the
adoption of space syntax techniques as tool to assess spatial entitlements in this research.

Access to a place or plot through a street is considered here as the first element of spatial
entitlement. It is not a surprise that regular street patterns or urban grids outlast centuries as
spatial solutions in new settlements. Hillier (1996:179) defines streets as the first powerful
theorem of urban engineering, due to the facility of enhancing efficiency of movement within a
city, which they create. The street system is transformed into a mechanism to generate contact,
by using each pedestrians origin-destination trip as an opportunity to create more encounters
and contact than the user previously intended. However, even urban grids are usually
differentiated according to locational variables. In colonial Brazilian cities, for instance, such
differentiation used to be caused by a major square where the cathedral and the city hall were
located, and by the port or road location, according to the case (Marx, 1991; Lemos, 1979).

The process of industrialisation and much faster urban growth has increased the importance of
location. Within developing country cities, such as Belm, the scarcity of means to provide the
same standards of infrastructure over the entire citys street system has caused further
concentration of activities and investment in central areas. It has strongly affected patterns of
density and land use, and accentuated qualitative differences between centre and periphery
(Lima, 2000). Conversely, the developed countrys city also tends to specialise its space by
creating discontinuity within its grid through enclaves or restrictions on multifunctionality,
favouring city sprawl, fragmentation and precinctisation.

In both cases the urban grid becomes a means to accelerate movement in order to overcome
size (Hillier, 1996:179), causing the replacement of loose movement over many alternative
combinations of routes by origin-destination movement. This shift has affected entitlement
conditions in terms of physical distance to desired public services and to sources of income,
such as domestic employment in middle and upper class household or unskilled jobs. Excessive
hierarchy within a city and/or urban fragmentation have also caused a reduction in the economic
potential generated by pedestrian movement along streets. Areas either sparsely occupied or
highly segregated prevent low income inhabitants from earning their living through street

5
vending, or from using the home as a shop or office, and increase dependency on public
transport by those who are already poorly served (Hillier et al.; 2000).

In spatial terms, the economic potential of pedestrian movement is historically related to the
structure of the urban grid and to its patterns of densities and land uses. This means that under
certain conditions of density and integration of a grid, things can happen that will not happen
elsewhere (Hillier, 1996: 170). The good space is the used space, the most frequent use of
space is movement, and movement is a by-product of the choices offered by a grid for getting
from one place to any other place (through all possible routes available by a citys street
combination). The informal use of space is strongly pedestrian-movement-related, as is the
sense of urban safety (ibid.:170). In Hilliers opinion the urban grid is the primary source of life in
cities, because of its capacity to favour mixed activities. He states that land use and density
follow movement in the grid, adapting to it and multiplying its effects (ibid.).

Conversely, in physical terms, structure can be a property opposed to order. Order is made up
of similar parts arranged through similar relations. It is best achieved when built up or imposed
all at once (ibid.:235). Deformed grids, which do not have enough repetition of elements and
relationships to create such an order, usually have a strong structure: spatial patterns that
neither can be seen at once nor are imposed at once. These grids are asynchronous both in
their genesis and in the way one experiences them (ibid.:235), although they are made
intelligible by the process of living, and, more importantly, moving in a town, according to Lynch
(1960), through landmarks and affective associations.

The informal settlements space is mostly asynchronous, and hardly makes sense to those who
do not live in it. This fact prompts its classification as a disordered space. Nevertheless, both
order and structure exist in the abstract space and in the built form, but order is natural to
form, or to what is built synchronously, and structure is natural to space, or to what is formed
asynchronously (trough independent and non-coordinated actions). With these differences in
mind, it is easier to understand why up-grade solutions to informally produced spaces should
not be compromised by formal standards and should be negotiated with the user community.
The existing structure must be considered and assessed to avoid arbitrary imposition of
simplifying order.

For this reason, observation of the relationship between local and global scales of accessibility,
and of the potential of movement within existing street layouts, is used here as a means of
assessing the condition of entitlement of informal settlement inhabitants (Box 1). Streets, like
classical legal entitlements, may be either obstructed or connected (rights are either suspended
or given) to the city grid, either restricting or improving access of users to desired activities. This
varies according to the social practices and governance policies adopted in the city over time.

6
 A brief presentation of Space Syntax basic concepts
Space syntax is an exploratory technique to find either problems or potential in the urban
structure, rather than to offer solutions or to give definite results (Hillier, 1998; quoted in Salheen
& Forsyth, 2001: 96). It evaluates the potential for physical accessibility and helps the
understanding of social patterns which arise in those spatial structures by using an axial map to
represent the topological relations of visibility and permeability of a space. The axial map is
drawn through the insertion of the fewest and longest straight lines of sight and access into all
circulation routes, wherein each line represents how far one can see and move from a space.
Space syntax assumes that pedestrians decide their route according to the minimum number of
changes of direction they must perform to move from one space to another (Hillier & Hanson,
1984; 1998).

Each change of direction is called a step, and the number of steps of a system defines its depth
or shallowness. Integration is the mathematical measurement of the relative depth or
shallowness of one line in relation to all others in the system. This is the most important
syntactic measurement, and can be calculated from a determined point within or outside the
system. When accessibility to all lines of the system is taken into account, this is called global
integration; but when the measurement is restricted to the access from one line to three or some
other established quantity of lines, it is called local integration. Global integration characterises
accessibility of the whole system, while local integration works with portions of the system,
being more related to pedestrian levels of accessibility than to the wider whole (Hillier &
Hanson, 1984; Hillier, 1996).

Space syntax benefits from the urban systems relationship between lines (streets and roads)
and built form to be reached through them. When there is correspondence between expansion
of streets and addition of built forms, line length is proportional to built occupation. Line length
also provides indications about connectivity (number of intersections of a line). The
incorporation of the topological measure of integration by the axial lines associates the lines
capacity to attract movement to the built configuration. Therefore, patterns of unequal attraction
within the urban system reflect inequalities in line length and connection (Hillier & Hanson,
1998).

The correlation between global and local measurements (e.g.: global integration versus local
integration, or control, or connectivity) is called the intelligibility of a system; it expresses the
quality of space in providing understanding to people about the global structure of its street
system from information received about the space they are in, or how far it is possible to capture
an idea about the whole from the experience of one of its parts (Hillier, 1989:63). Values of
intelligibility tending to 0 reflect a labyrinthine structure, and values tending to 1 express
excessive regularity. Typical urban areas present a correlation around 0.45 (Holanda, 2002).

Space as provision

Space as provision develops through the transformation of natural into built environment,
through human agency and according to available resources. The urbanisation of natural space
was speeded up in Brazil after the 1950s, through the incorporation of capitalist relations of
production into the process of urbanisation and city expansion (Santos, 1980). Landowners,
entrepreneurs, funders, big companies that buy and develop land, all had particular interests in
this process, which led to the repetition into the cities of the existing unbalanced relations
between city and countryside and between different regions of the country. This resulted in
abundant service provision in central areas and scarcity of them in peripheral ones. The best

7
served areas became contested for by the wealthiest activities, and occupied by high rise
buildings. Prices became a function of accessibility to the centre, and low income housing was
feasible only on the outskirts of the city where urban facilities were minimal, and prompt
solutions were based on two distinct alternatives, informal settlements and official settlements.
Box 2 presents explanations about how commodification of land happens in Brazilian cities.

As the quantity and quality of housing officially provided for the poor is insufficient and more
expensive than it should be, the poor have produced space to live in, conscious of the
unavoidable struggles to be faced in order to upgrade this space over time. It is a collectively
and spontaneously produced space and needs to be assessed through approaches based on
the idea of change. Urban morphological approaches based on empirical data (Cannigia &
Maffei, 1995; Conzen, 1969), which have explained vernacular occupations in contexts with
different levels of material infrastructure, may offer suitable tools to discover which elements or
characteristics of the production (or provision) of informal settlement space help enhancement
of inhabitants life chances.

The investigation of the evolution of informal settlements urban form over the time span of its
formation may help understanding of how and why entitlements were given or withdrawn. As
defined in the theoretical concept of life chances, provision is affected by supply conditions and
is dependent on economic growth, so physical provision may qualify spatial entitlement, through
the qualitative and quantitative diversity of access given to the former.

Urban morphology and footprints of informal settlements

According to urban morphologists, cities develop themselves by changing their physical form
continuously (Kropf, 2001) through the action of different agents and producers of space. The
primary determinant of urban occupation is the site, but urban settlements are initially
materialised through streets that connect two points: a pole (building or place that works as an
urban generator) and an anti-pole. Through non-planned urbanisation, streets gradually define
block shapes, according to the evolution of occupation along them. Streets also are historically
differentiated; for instance, occupation starts in a primary street, prompted by a focus of interest;
when a second street (street of implementation) intercepts the first, the urban fabric starts to be
generated. Afterwards, streets of union connect streets of implementation (Caniggia & Maffei,
1995) (Fig 1). These streets may be differentiated by size and shape of plot, each presenting
different levels of density, creating a hierarchy and a potential for different land uses (McGlynn,
2001, tutorial). The intensity of occupation usually increases with the proximity of the main living
focus in the settlement, towards which more intense fluxes are oriented (Dos Santos, 1988).

Fig 1. Models of urban fabric formation. A primary streets; B streets of implementation; C1

streets of union generated previously to implementation streets extension; C2 streets of union
generated after implementation streets extension; D street of up-grade (Cannigia & Maffei,
1995:88)

Blocks are progressively defined by each streets series of plots, which are occupied in parallel
(forming a route rather than defining sides of a block). Series of plots facing each other usually

8
have a synchronous occupation, reinforcing the importance of routes over blocks in
morphological terms (Caniggia & Maffei, 1995). In regular conditions of occupation, the site is
considered the most permanent of all physical elements (Brand, 1994), but in the case of
flooded informal settlements it is still in a process of generation, coming after the street pattern
(defined by elevated wooden walkways, called in Belm estivas). In these cases, the site is the
determinant of settlement location, to the extent that sites with disadvantaged physical
conditions (flooded plains, hills, hazardous areas) have occupation postponed as much as
possible, and often are the only areas available to the poor. The sites physical condition also
determines intensity of occupation (density) and level of upgrade investment (infrastructure
provision) within the context of poverty.

Urban morphology often organises empirical data through cycles, periods and series.
Morphological periods, for instance, are defined by Conzen (1969) as any period in the cultural
history of an area which creates distinctive material forms in the cultural landscape to suit the
particular socio-economic needs of its society. These forms survive in varying degree as
residual features (ibid.:127). These periods may correspond to cycles of building development
according to conditions of land occupation and socio-economic demands and may produce belt-
like zones of peripheral mixed land uses when growth becomes stationary. This research
approach searched for a correspondence between morphological periods and consolidation
cycles, in order to identify morphological clues related to different stages of consolidation.

However, the whole process of space generation is just part of spatial provision; all spaces
created are expected to synchronously receive infrastructure according to existing demand and
affordable technical standards. This does not happen in informal settlements, due to the usual
increase of people and plot coverage density at a higher rate than infrastructure provision,
aggravated by the producers socio-economic constraints.

Density and infrastructure

The informal settlements space is a means of survival. Inhabitants manage spaces contribution
to the satisfaction of basic needs. When plots are bigger, and density is lower, subsistence
cultures and orchards help livelihood, and creation of own water and sanitation systems is more
likely to be possible (Madaleno, 2000; Choguill & Choguill,1996). When land becomes scarce,
often due to lack of accessibility from non built-upon land, existing plots are subdivided and
density increases. The possibility it creates of per capita costs reduction in infrastructure
provision is undermined by the technological standards usually proposed to these densities.
This results in delays or absence of investments due to the financial constraints of inhabitants,
who continue to live in, not only flooded, but also overcrowded areas. This begins a vicious
cycle of poverty, environmental degradation and health problems. This often jeopardises
access to important assets such as clean water, and to sanitation and rubbish collection
(Drakakis-Smith, 2000; Vieira, 1997; Acioly & Davidson, 1998). Physical (and also social)
infrastructure provision is seen as a long-term prospect and is dependent on the prevailing
political and technical paradigms of urban governance (Choguill & Choguill, 1996).

Users are expected to pay for infrastructure provided by official or private agencies, and its cost
depends on established standards. Informal settlements try to benefit from accessibility to social
infrastructure available nearby, but require a proper solution to their lack of physical
infrastructure. Intermediate technology has proved to be more suitable to inhabitants income
and environmental and health objectives in many circumstances, but has not accomplished
objectives in Belm up to the present, due to lack of inhabitants understanding of alternative

9
systems conception, lack of maintenance, and poor site conditions (Mendes, 2003). Further
research is still to be carried out to develop cheaper infrastructure technologies suitable to
Belms flood plains.

Conversely, density, within an informal context, depends on household size, land availability,
settlement layout, size and shape of plots, and typology of buildings (other important variables,
applicable to formal contexts, are legislation and level of infrastructure available) (Acioly &
Davidson, 1998:23). Higher densities are not a problem in themselves, as very dense cities
(such as Amsterdam) cope very well with high densities. There is a problem when high density
constrains sustainability, which may happen when it is not followed by compatible infrastructure
solutions. It can cause overcrowding, difficult access in an emergency and conflict with local
climate requirements (as in hot and humid equatorial areas) (Williams et al.,2000; Schiller &
Evans, 2000).

From the perspective of Belms case study, priority should be given to the brown agenda, due
to the high likelihood of water contamination and unhealthy living conditions in all the case study
areas. Conversely, the process of site transformation raises issues watched by the green
agenda (landfill of flood plains, removal of vegetation). The object of interest here are the
impacts of solution on informal settlements space, rather than the particular solution of water
and sanitation provision. Moreover, healthy and safe environments on a city scale need the
support of national policies, and within benefited communities any program needs to be
complemented by hygiene education. Within the complexity of the informal settlement
consolidation process, more attention could be given to how things are done rather than to
results, according the recommendations of Williams et al. (2000) about how to achieve
sustainable urban form.

Space and ligatures

According to the concept of life chances, ligatures are the linkages between entitlements and
provision (Dahrendorf; 1979, 1988). They are defined as producing human motivation, shaped
by traditional, social, cultural, psychological and other values. Ligatures give sense to choices
between eligible alternatives, and have shaped space in different ways all over the world. The
research approach developed in this chapter searches for a spatial dimension to each element
of the concept of life chances; as the most abstract of all, ligatures is assessed here through
street life, because of the latters capacity to express relationships between space and its users,
and therefore to unveil human motivations and profiles.

At present, many authors (Sennet, 1993; Soja, 2000; Castells, 2000) offer a picture of a world-
wide trend which will end urban life as public life; structural inequalities have generated social
unease and violence, and often lead to measures of self-protection by those who are at the top
of societies. This is clearly exemplified by the extreme case of Californias increasing numbers
of gated communities, which expect to be protected against the minorities of poor, Hispanics,
blacks, and others disliked by the mainstream society.

Levitas (1991:228) claims that life in the streets, like any cultural artefact, reflects the adaptation
of modern culture to the environment, and therefore changes in the technological and
sociopolitical configuration of our societies are keys to understanding the present decline of
street life. For instance, industrialisation and centralisation impacted on daily social life in urban
spaces through automobiles, television and definition of new economic scales that made small
personal business unprofitable and favoured the growth of impersonal supermarkets and

10
shopping centres. These made division between rich and poor more extreme and increased
social tension and violence on streets.

The evolution of street life in Western cities shows how streets express forms of social
organisation by creating systems of barriers and permeability that help to regulate social
interaction among social groups. Streets previously worked as centres of information, but
became progressively specialised according to the rhythm of progress in technology and
capitalism. The identification of streets with social class was expressed through their physical
amenities and rhythm of occupation. The emergence of a global economy, and transformation of
cities into nodes of the global network, has overemphasised the circulation and symbolic
dimensions of streets. These favour centralised power and control, and indicate a trend to
homogenisation of public spaces that serves the interest of upper and middle classes, but
neglects the basic need of children and elders for attachment to a primary space that should be
stable and predictable (Levitas, 1991; Weber, 1964).

Since Vitruviuss time, architects and designers have tended to have a platonic faith that
symmetry and harmonic proportions would ensure perfection in space. This has created a
tradition observed in present design and planning, favouring aesthetic ideals and a belief that
forms have an impact on peoples social behaviour and happiness (Levitas, 1991: 225).
However, Rapoports (1969) findings showed that such rules do not create the effects which
designers credit to them. In the rare cases in which design appeared to have a strong effect on
behaviour and user satisfaction, there were social and psychological determinants in addition to
physical ones.

Anthropologists suggest that ecological approaches (able to integrate livelihood systems,

architecture, social behaviour, etc.) would be more effective in investigating the impact of each
single aspect on inhabitants lives, without considering any of them determinant upon the others
(Douglas, 1970, cited in Levitas, 1991). The anthropological approach stresses participant
observation, and immersion of the researcher in the culture he is studying in order to acquire a
better understanding of inhabitants attitudes and the interdependence of material and non
material aspects of culture (Levitas, 1991: 227).

To clarify the contribution of space to human interaction, Granovetter (1982; quoted in Hillier,
1996) highlights spaces ability to generate weak ties, or ties created by occasional interaction.
Weak ties act as bridges between more homogeneous groups formed by ties of blood and duty
(strong ties) and are considered by him the key to urbanity. The weak and strong ties are the
social equivalent of the local and global scales of space, which must be balanced to create a
well-integrated environment. Hence, space offers the middle ground between the local
community and the transpatial network that is more or less independent of space (Hillier,
1996:257).

Other studies of the effects of social and physical distance on human interaction (Knox & Pinch,
2000:222) have shown that instrumental interactions (caused by professional and political
interests) tend to be less dependent on distance (more transpatial), than instrumental interaction
to achieve a common goal (e.g. to organise a party or to avoid the closure of a school), which
presents more interrelation between physical and social distance. Voluntary associations (as the
poors networks of mutual support may be considered), based on class values and lifestyle
(established to sustain social relationships) are also strongly correlated with space when there is
correspondence of social and residential segregation.

11
In the 1990 Poverty World Development Report (World Bank, 1990; quoted in Moser, 1998) the
reciprocity between households of a poor community, based on social ties, is seen as an asset
to overcome poverty, as important as are health, education or house ownership. Box 2 presents
findings of previous research about the role of streets in social networks in American slums; it
illustrates street life of a kind much closer to informal areas in Brazil than non-urban
communities in Asia and Africa would be. This data indicates that the time of consolidation and
level of segregation of a community play an important role in determining the quality of the
social network formed in them, basically in their streets.

In this paper, it is assumed that there is correspondence between communities and space in
informal settlements; the investigation of ligatures is done through observation of streets, to
learn about the poors social contact with others in their neighbourhood by observing the use
they make of space, or their environmental behaviour (Zeisel, 1981). From the spatial
perspective, and in formal contexts, quality of social relationships are associated with the
existing gradients between public and private spaces (Ford, 2000). The ideal sequence would
be public, semipublic, semiprivate and private spaces. The first is represented by the busy street
that should lead to the communitys internal streets and to the space between buildings, and
then to a buildings interior. This hierarchy of streets should offer to inhabitants the possibility of
enjoying territory and avoiding social and psychological problems in groups highly dependent on
space, such as young children, who need to form locally-based friendships, and those who
cannot afford the benefits of improved mobility and communication (Knox & Pinch, 2000;223).

Sidewalks, porches or verandas, walkways, steps, fences, among other spatial elements, are
the physical amenities available to distinguish the different gradients between public and private
spaces. In this way, demarcations of space or their absence may help the creation of bridges
between communities, to ease social exclusion, and to favour co-operation rather than violence
and competition within cities (Castels, 2000). In opposition to the Californian example, Castells
cites the case of Barcelonas spatial interventions to promote peripheral centrality, an attempt to
recreate the meaningful structure of the city, in order to prevent neo-tribalism physically, and
provide for children a safe, open, fun and dynamic city, to allow them to grow up happily
(ibid.:122).

12
The role of streets in slums
Studies focusing on American slums as a whole and on juvenile gangs in particular indicate that
streets are extremely important in slums. They provide primary references for their inhabitants
and a sense of belonging and cohesion; they are also places to relieve the constraints of
domestic life (Suttles, 1968; Yablonsky, 1962, cites in Levitas, 1991). Streets are the places of
interaction of different ages, genders, and groups; they are also places of gossip and
interpretative observation. They are the means to communication, since they provide
opportunity for interaction at the same time as they protect inhabitants scant privacy (Suttles,
1968:77).

Privacy in slums is scarce even inside homes, due to crowding, large families, and poor
construction that always favours intrusion. In these environments, social life becomes so
informal that any domestic exchange might enforce unpredictable exposures or confrontations
that are not easily moved back from (ibid.:77-78). The street offers an escape to slum dwellers;
it provides excitement and unscheduled experiences, and training in a relatively rigid system of
relationships that provides a few techniques to encourage participation in the larger society;
informal network gossip is an effective means of information about the outside world (Liebow,
1968; Ulf Hannerz, 1969; Herbert Gans, 1962; Whyte, 1955; Lee Rainater, 1966; cited in
Levitas, 1991:235).

Suttles (1968) found that there seems to be no standard of morality in street slums; therefore
there are no general norms to be taken as a reference to evaluate individuals. In these areas,
groups are differentiated by background and patterns of behaviour, and information is used
most of all to maintain personal relationships, not a sense of community. Only when there is an
external threat information is used to benefit the community. Groups share and keep their
knowledge, and street gossip is an effective but quantitatively limited means to gain knowledge
of inhabitants personal characters. This system of information limits inhabitants access to
additional information from outside; this might result in the creation of stereotypes by inhabitants
about the larger society, and difficulties in dealing with it (the wider society) realistically. (Suttles,
1968, cited in Levitas, 1991:234).

However, streets are still the best environment in which inhabitants can adjust and adapt
themselves to a larger unit, and to experiment with different life-styles. Men look for a place they
can call their own, since the domestic space is often considered a female world. In slums with
higher levels of integration between groups and a longer time of consolidation, streets are used
in a less segregated way. Age, sex, and social roles are combined with ties of kinship, work,
religion to the point that Herbert Gans (1962, cited in Levitas, 1991) decided to call them The
Urban Villagers.

The role of streets varies according to the level of settlement segregation, but has the potential
to offer to slum inhabitants opportunity of socialisation in expressive rather than instrumental
terms, and provides them with education and security, although that security imposes the price
of limiting flexibility and the extent of social networks (Levitas, 1991: 235).

The case study

The starting point to investigate the informal settlements consolidation process in Belm, was to
select two basins, one located within the city centre (called Tucunduba) and another in the
expansion area (called Paracuri). Both are characterized by gradual occupation of flooded
plains and by a originally peripheral condition. Tucundubas occupation began during the early

13
1960s and Paracuris occupation by late 1970s. Selection of case study areas within the basins
was conditioned to factors such as tenure ownership, density, intensity of upgrading, and
inhabitants social-economic profiles. This selection has allowed the association of location and
time of origin, and the observation of distinct stages of consolidation through the different case
study areas configurations (Fig 2).

Within the framework devised, location expresses the possibilities found by inhabitants of
overcoming distance, given by socio-spatial characteristics of the city. Configuration shows the
physical arrangements between spatial elements in a given moment, it is considered as a static
snapshot of a process of change which occurs over time, through upgrade actions practised by
inhabitants and other agents. Timescale introduces the time dimension that allows observation
of the successive configurations developed over a process of consolidation and of changes in
the relationship between space and agents who are producers of space (McGlynn, tutorial
2002b).

Fig 1. Left: Map of Belm. Doted line maks out the city centre, and continuous line the basins
selected. 1- Tucunduba basin; 2 Paracuri basin. Fonte: Cardoso, 2002; CELPA, 1998. Right:
Belms land form, the palest color represents waterlogged areas. Fonte: JICA, 1991.

About entitlements

Location and configuration were investigated through comparisons between case study areas
syntactic measurements; timescale was observed through the time lag between the case study

14
areas time of origin and comparisons between maps that isolate different moments of
settlement evolution in both basins. The three variables were checked against socio-economic
variables obtained from questionnaires, applied during a fieldwork, to refine the spatial
evidences.

From the outset, global integration levels indicate that both areas are segregated in relation to
the whole city system; however, Tucunduba basin has the advantage of being closer to the
citys most integrated areas, where diversity of social groups and commercial activities create
opportunity for casual jobs and informal street vending (common sources of income of the poor).

Observation of global and local integration measurements of the basins street system in three
moments in time (1977, 1986 and 1998) showed that over two decades there was a
metamorphosis in Tucundubas street system. This resulted in the creation of a balanced and
well-structured vernacular system after the introduction of new infill grids and upgrade actions,
consisting of drainage and street regularization (Fig 3). In Paracuri basin the street system has
only recently evolved; the street system formed by the informal settlements are isolated from
each other and still peripheral to the formal grid.

Observation of evolution of street depth over the same period showed that there is a two step
grid more or less constant in the city centre, which has received commercial activities, services
and social infrastructure (schools, health centre). However the maintenance of original
circumstances of some streets over this length of time (the blue/violet lines), despite the general
positive changes, showed that time needs proximity to most integrated lines to lead to decrease
of depth, and enhancements in integration (or accessibility conditions) (Fig. 4). This was
confirmed, but with a weaker intensity, in the expansion area.

5.6270 5.6270 5.6270

0.2109 0.2109 0.2109

Measure =3 Measure =3 Measure =3

Fig 3. Tucunduba basins local integration maps (1977,1986, 1998). Legends have the same
range of colours to allow visual comparison between maps. Colour range presents descending

15
order of local integration values; red indicates the most integrated lines; deep blue indicates the
most segregated lines.
Sources: CODEM 1977, 1986, 1998; CELPA/CODEM, 1998.

-1.0000
-1.0000

-1.0000

-9.0000
-10.0000
Measure =0
-9.0000 Measure =0
Measure =0

Fig 4. Tucunduba basins point depth maps (1977,1986, 1998), drawn from the two most
important streets in the system (the line northeastwards is the second most integrated in the
city, and the line southwards is the locally most integrated of the system). Both lines are red and
marked by dots. Colour range presents descending order of depth.
Source: CODEM 1977, 1986, 1998;CODEM/CELPA 1998.

Social variables were introduced at this stage of analysis to check the achievement of life
chances against the potential for integration or segregation of space. Modes for schooling
years against local integration showed a distinct differentiation of levels of education in
Tucunduba basin according to levels of local integration, and similar trends in the expansion
area. Time of journey from house to school and means of transport from house to school
qualified the results for each case study area, showing that it takes longer to go to school from
those areas found to be more segregated within the city centre and in the expansion area in
general. This means either the existence of longer distances between house and school, or a
search for better standards of education outside the district of origin, and reminds us that
homogeneity of socio-economic conditions delays qualitative improvements in social
infrastructure.

Cross-tabulation between highest income per household and local integration showed that work
with a fixed income is more frequent in Tucunduba basin; upper level income is associated with
intermediate levels of local integration, and low income is spread over all the levels of local
integration (although slightly concentrated in low intermediate levels). In Paracuri basin, most of
the economically active population has uncertain/unpredictable income; low income households
are present at all over the system, and upper income people are in the low intermediate levels
of local integration. Although upper income levels present a clearer pattern in relation to local

16
integration than other income groups, it was not possible to make conclusive statements about
the contribution of space to access to income, based only on the relationship between these
variables.

In search of further evidences, cross-tabulation between household size and local integration
showed that in Tucunduba basin five to nine member households are more frequent in
intermediate levels of local integration, while ten inhabitant upwards households happen at all
over the system. In Paracuri basin three to four member households are located in most
accessible streets, while nine to ten member households are located in intermediate levels of
integration (and accessibility). It was an indication that household size does help in the
achievement of better location, but the number of breadwinners needs to be considered in
assessing the extent of that achievement.

Cross-tabulation between the ratio of workers to non-workers and local integration confirmed
that number of breadwinners is important. In Tucunduba basin, when at least one third of
household inhabitants work, the large families are more likely to be settled in the higher levels of
local integration, while when ratios are inverted, large families are more frequent in the lowest
levels of local integration. In Paracuri basin there are cases of total absence of income, and
cases of households with workers to non-worker ratios of 0.1 up to 0.25 are twice as high as in
the city centre.

The period of settlement of the surveyed household showed that, in Tucunduba, settlement was
progressive along the upper intermediate levels of global integration. The period of most intense
settlement was the 1970s; since then the rate has progressively decreased. The lowest levels of
local integration accompany pre1985 and post 1995 settlement, showing again that time alone
cannot provide enhancement of local integration levels. Moreover, there are newcomers settling
in the most segregated streets and in the intermediate levels of local and global integration
where physical improvements were made. As much as gentrification happens in improved
streets, poor newcomers are still settling in segregated streets (usually narrow, flooded and
short); this highlights the appeal the basin has to the poor and possible advantages of a slow
process of transformation.

The results related to Tucunduba basin confirm that the location within the city makes an
important contribution to inhabitants life chances. However, a great deal of this contribution
must be associated with configuration, which is more robust in the city centre than in the
expansion area. Configuration is responsible for the local dimension of location and interferes
with the contribution of time to the process of evolution and consolidation of the case study
areas. The association of the three variables creates a responsive space, able to meet the
requirements of different social profiles in the city centre. In the expansion area, location affects
access to income negatively for the poorest, but the relationship of space with other social
variables seems to follow patterns similar to those found in the city centre.

Inhabitants socio-economic data allowed some considerations about how they have benefited
from the potential delivered by space. In terms of education it was observed that continuation of
education has been higher in the city centre, and even more, in the city centre it is easier to
young adults to resume school attendance. Levels of education are usually low, but the
proximity of city centre case study areas to the public university has allowed cases of original
inhabitants second generation access to higher education. In general terms either depth or
local integration have predicted achievements in education. In the expansion area, local

17
integration predicts achievements in education, with similar prospects for formal and informal,
but more consolidated areas; recent settlements offer worst prospects.

Regarding to income, the most consolidates area shows higher levels of self-employment
among adults (professional and unskilled), and a lower level of stated unemployment (perhaps
some were identified as students). This may confirm potential advantages the space delivers to
informal activities, which benefit from the streets accessibility potential (on sidewalks or in front
of houses). In another case study, less consolidate and accessible than the oldest, most adults
are employed and rates of stated unemployment are higher, with less self-employment. This
suggests lower possibilities of providing opportunities through street accessibility in this area
than in the former one.

In the most segregated case study area within the city centre, the profile of teenagers
occupation is different from that of the previous areas. Work starts early in life, and the majority
of workers do not have professional skills; however, levels of stated unemployment are low.
This might mean that, within this area, inhabitants skills are inappropriate to the citys
employment standards, due to their origin (higher proportion of migrants). Nevertheless, they
somehow cope with their spatial limitations, either going to work (usually vending) on streets
outside their living space or performing casual jobs.

Rates of stated unemployment are higher in the expansion area than in the city centre, and
among workers there is a prevalence of low skilled self-employment. Less than half of
teenagers are students in the most recent settlement; and mature adults are not retired, which
means that the household cannot rely on pensions, they are either self-employed or
unemployed. This might mean the perpetuation of low skilled jobs, and indicates a clear
difference of achievements between the population in the city centre and in the expansion area.

About provision
The space of case study areas is a vernacular space, produced according to its inhabitants
economic constraints, and required an appropriate method to describe and analyse its patterns
of production. Site configuration has conditioned occupation from the existing grid towards a
river (Tucunduba River within the city centre and Paracuri River in the expansion area), which
are used as routes of access. More consolidated settlements present at least two important
edges in their process of occupation. In the expansion area there is only one important edge in
each settlement. The application of Cannigias & Maffeis (1995) typology resulted in a
classification of streets according to their process of generation and evolution, and the
identification of peculiar patterns of street layout in each basin.

The street layout in Tucunduba basin, already identified as grid-oriented by syntactic

measurements, showed itself subdivided in different scales according to the arrangements of
the street system created through subdivision of the fat original blocks simultaneously with
density increase. In Paracuri basin, the route-oriented street layout offered a diverse procedure
for expansion, extending implementation streets over the unoccupied territory, and creating
narrower blocks than the original ones in the city centre (Fig 5).

Cross-tabulation between type of street (primary, implementation, union, upgrade) and depth
clarified the hierarchy of grids and showed that the process of centrality is repeated on
decreasing scales in Tucunduba basin. Even among the most segregated streets, there are
arrangements to interconnect grids (embedding new grids) which follow the same principle of
centrality presented by the whole system. These are too segregated to be identified only

18
through the maps of depth presented in the previous section. A similar process occurs in
Paracuri basin, despite the differences of scale and intensity.

Cross-tabulation between type of street and street conditions showed that the generation of the
grid hierarchy follows a gradual process of street improvement (drainage, landfill and pavement
occur first in the streets with higher potential for accessibility). Cross-tabulation between depth
and street conditions showed that the first predicts the latter in the city centre. There is a
correspondence between spatial and socio-economic differentiation expressed by the distinct
amounts of investment noticeable over different depths. In Paracuri basin the grids are still too
loose to generate a strong differentiation of streets in the case study areas.

Cross-tabulation between street type and street width supported the grid hierarchy in
Tucunduba basin and the dominance of streets in Paracuri. Street condition is fundamental to
the gradation between private (the internal space) and public space (the space shared by
inhabitants and outsiders) in the city centre, while in the extension area, wider streets allow plot
and building features to contribute to open space differentiation. This sequence of findings
expresses how the condition of scarcity determines the land market in Belm. Globally, location
contributes to provision through the status it gives to each area within the city land market.
Within a third world city this is associated with city growth; always when a new settlement is
created in the periphery, the value of existing land increases towards the city centre. Central
areas become more sought after and scarce, most valuable and prone to receive more
investment.

Timescale defines street hierarchy and change in street status, through extension of occupation
and infrastructure provision. Streets that extend settlements usually have worst infrastructure
conditions. Location and timescale indicate the tendency of settlement aggregation. The oldest
settlements, located in the city centre, form bigger areas over time, which in turn originate
districts. New settlements lack secondary links.

Increase in extension of building coverage and increase of built density are negatively
associated with the citys climatic conditions, and with infrastructure solutions affordable by
inhabitants, increasing health problems (and disadvantaged work performance) and the
perpetuation of poverty among inhabitants. Nevertheless, higher densities are positively
associated with social infrastructure provision when present in extensive areas.

Moreover approximation between the present levels of density in Tucunduba and Paracuri basin
(Santa Cruz and Arthur Bernardes) shows that the influence of time is decisive in global terms.
It is incorporated through the evolution of the land market in the city as a whole, and changes in
the action of agents (e.g.: the settlements in the expansion area are already privileged locations
compared to others located on the outskirts of the other municipalities that constitute Belm
Metropolitan Region).

Further comparisons showed that tenure regularisation helps to keep density lower, due to clear
definition of plot and block limits, preventing further invasions arising from the attractiveness of
accessibility. Up to certain levels of density, the space analysed showed itself extremely well
structured, but unhealthy, due to lack of infrastructure.

When inhabitants were inquired about provision it was noticed that purchase was the most
usual form of access to housing, plots could be sold by an informal developer or by a previous
owner. Invasion itself was found only in the newest cases and where the process of tenure

19
ownership was not in progress. Inhabitant perception of provision is not very diverse from
settlement to settlement. Everywhere there is a strong perception of street change. However,
changes were said to stronger where there was official upgrade actions have happened (usually
sanitation related), regardless to accessibility potential. Transformations in the most
consolidated case study area have stagnated after preliminary drainage and landfill, while in
another case study areas drainage resulted in good paving of longer streets.

Inhabitants always see themselves as active agents in a sites initial physical transformations
and either the government or politicians as agents of infrastructure provision who come up in a
second moment. Responses also indicate that interventions should happen during the early
stages of consolidation in order to make changes in street alignment feasible and qualitatively
superior to what has been delivered by current upgrade actions.

Inhabitant also show a prevalent willingness to trade-off house ownership against environmental
conditions. All efforts are made to avoid compromising their finances by rent payment,
independently of the type of previous housing. The informal settlements surveyed have mixed
origins, and inhabitants have shown they are aware of how to incorporate value into their plots
and houses, following the same rules as the formal land market.

Inhabitants slowly landfill and fence plots and improve houses according to money availability.
At least 60% of all surveyed households have adapted their house. House adaptation is also
seen as an investment, since in 40.5 % of surveyed houses, present inhabitants are not the first.
It was also found that gentrification happens in segregated and integrated streets, but seems to
be more intense in previously upgraded areas. These perceptions are strengthened by the type
of enlargements made to the houses. The most frequent change is a back addition to provide
more rooms (either for the household or to rent), followed by additions in front of the house,
usually to provide space for a shop. A second storey and the building of an independent house
close to the original one are the other usual additions. These provide more flexibility but require
more investment.

The perception of housing provision as an asset to income generation is stronger in the

expansion area, where the house provides place for the main household economic activities.
This is often done by the better-off, those who own a shop or have a workshop and are living
amidst the poor. In the city centre also, similar cases of shops and workshops for income
generation based at the home are spread all over the area on a smaller scale, and perceived
more as complementary to other income sources; in both cases configuration plays an
important role, through creating street accessibility and a favourable relationship between house
and public space.

The amount of investment in streets and housing should be understood as critical to the
contribution of provision to inhabitants life chances. Configuration, through depth, seems to play
an important role in this, defining the potential for exploitation of the house as an asset, ranging
from its providing freedom from rent, in a segregated street close to well-served areas, to it
enabling the establishment of home-based income generation, where occupants benefit from
better accessibility.

Access to surrounding areas on foot was an advantage of living in the city centre case study
areas, as access by bicycle was an advantage in the expansion area. In the city centre, street
upgrade has improved accessibility for the most mobility-disadvantaged groups, such as ill
people, the disabled and pregnant women.

20
All informal settlements have some degree of disadvantage regarding public health, due to lack
of drainage, sanitation and waste collection; all problems caused by insufficient or absent
infrastructure. As a form of compensation, land uses in a settlements vicinity are considered
very important to inhabitants. This is clearly observed in inhabitants concern about easy access
to hospitals, emphasising the importance of health care to compensate for their vulnerability to
diseases generated by their environmental conditions.

About ligatures

Street life was taken as an expression of inhabitants ligatures in relation to space. The starting
point was the potential offered by case study areas urban form to inhabitants social interaction.
The relationship between streets, block, plots and buildings (e.g.: houses opening straight into
streets, setbacks which work as alleys to inner block occupation, attached typologies) creates a
system of barriers and permeabilities, gradually differentiated by potentials and hierarchies
previously explained. This system is still flexible enough to allow diverse patterns of behaviour,
according to prevalent social codes; outsiders have access all over the areas, but under
progressive control of inhabitants. The intensity of control distinguishes the most segregated
areas from the most integrated and affects commercial activities, which are more intense in the
city centre and limited to one street or a few street junctions in the expansion area.

In the most consolidated areas there is a stronger interaction between inhabitants and outsiders
and both contribute to street life, either along streets or concentrated at street junctions. In the
less consolidated areas interaction is more restricted to inhabitants along estivas, resulting in
social segregation and restricted availability of references for urban codes of behaviour. In
intermediary stages, streets work as extensions of the house space, and are places of
information exchange and shared activities, especially related to leisure or informal activities.

Inhabitants, as users of open spaces, can be broadly classified into two profiles. One is of
aspirer, who are looking forward to improvements, and aim to control unwanted interaction
through distance, pursuing a home-centred life style and practising co-operation as courtesy;
they are either children of original inhabitants or better-off new comers. The other is of
conformer who rely on strong spatialsocial networks, and present a street-oriented life style,
accepting present conditions either by managing constraints with flexibility and creativity, or by
being indifferent and careless. Attitudes might be associated with conditions of access to
income, length of time in the settlement and previous living conditions.

From observation of the use inhabitants make of space, four main categories of activity were
identified: work, consumption, entertainment and caring (either of people or space), combined in
diverse intensity and frequency. In all areas, streets are fundamental to inhabitants lives.
Through their configuration they provide economic opportunities, and social references to them,
through the interaction among different internal groups and between inhabitants and outsiders.
However, most activities performed on the streets lack physical amenities and conflict with
traffic especially car traffic. Outsiders who usually drive through areas located in the city centre
have no wish to take part in the life that happens there, to the disadvantage of inhabitants who
could benefit from such contact. In the expansion area, the most consolidated case study area
is a quiet and stable place, and new settlements are still struggling to have proper streets.

After all, trade-offs, motivations, values and networks of case study areas inhabitants were
investigated, in order to present the ligatures that make their lives feasible in the environment

21
where they live. The trade-off between physical conditions and location was confirmed in all
originally informal areas, except in expansion areas most consolidated one (where there were
higher levels of gentrification). In all areas surveyed, location was pointed to as either important
or very important. Price was considered very important or important in all areas.

Inhabitants motivation was primarily economic; house ownership is a very important step in
overcoming poverty, and re-sale is always a possibility if people can afford something better
than what they have. However, it is not easy to sell a house, especially when the prices are high
compared to the formal market, as is sometimes the case (speculative behaviour among sellers
and buyers are raised always that there is expectation of upgrade).

Responses to the questions about the house were very similar to those given about plots.
Inhabitants trade-off a sites physical condition, plot and house size and flexibility for a lower
price and better location. However, among the poorest, need is what actually controls plot and
house subdivision, and is the reason for the generation of cramped housing in the best
locations.

The assessment of physical qualities is diverse. In all settlements there is a perception of being
in a good location (mainly due to the proximity to an important avenue); although inhabitants do
not perceive the trade-off between location, house ownership, price and physical conditions
equally. Those who were initially poorest are the happiest; after years of struggles, inhabitants
do not want to move, but they want a better place for their children in the future. There is also a
willingness to welcome benefits more than to embrace obligations, expressed through the
attempt to avoid tax payments.

Perception of safety is worse in the city centre than in the periphery, and is associated with
mobility of population and increase in density, and is especially strong in the less consolidated
case study area. In the expansion area, the enhancement of connections between informal and
formal settlements and the settling of newcomers improved original safety conditions. However,
in the newest settlements the strong internal control and the unpredictable physical conditions
are still constraints on outsiders safety. Still in the expansion area, there is a stronger need to
create a limit to intimacy with neighbours next door, perhaps due to usual extension of the
household space into the street space, and the need for ensure solidarity without intrusion.

The sense of community is not evenly developed. Participation in formal communitarian

associations is stronger in the city centre (strongly related to womens action); besides that, only
the church is able to congregate inhabitants in all areas, the Catholic in the most consolidated
areas and the Protestant in the poorest ones. Political affiliation is rare, since political bargains
were usually more effective to bring about changes than engagement with a political party.
Socialisation happens most informally, through front door contact, sports practice, and children
playing, and is less usual where spatial conditions are precarious. But mutual support is usually
related to strong bonds of friendship, created between old neighbours or restricted to relatives.

Final considerations

The occupation of Belms flooded plains located in the city centre, was a strategy for housing
provision by those who were not entitled to formal housing. Entitlements and provision were
poor, but in a context of general scarcity there was a potential for access to urban life and to
housing. Evidence shows that choices made by invasion areas long-term inhabitants were

22
reasoned with regard to the type and amount of resources available, and allowed creation of a
fragile balance between people and space. Originally, the poor created informal settlements by
building a flexible space, carefully positioned in the city, which became extremely compact over
time, up to the point which made improvements feasible.

The initial limitations of the poor to entitlements led them to exploit the citys natural environment
and their social networks to create provision that would never have been achieved otherwise.
By arousing other agents interests, the poor consolidated the space they created and also
found a way to take part in the land market, profiting from commodification. Weak entitlements
and provision were enhanced through a strong sense of purpose, consciousness of trade-offs,
and strong ties to socio-spatial networks. From a top down perspective, the achievements might
appear unremarkable if formal standards are taken as references. However, from a bottom-up
perspective, it is possible to see that the poor have done their best to be integrated into the
formal city, creating a space extremely responsive to their needs, which if receiving proper
improvements, provides inhabitants with dignity and better prospects, and to the city, a long-
term asset.

From the exposed it is possible to conclude that the provision only approach to informal
settlements upgrade is insufficient to guide the action of other agents (government, funding
agencies, building enterprises, politicians) towards the maximum benefit of original inhabitants.
A bottom-up approach, built upon investigation of inhabitants socio-spatial strategies and
perceptions of their environment brought about intrinsic relationships between space and
inhabitants, has shown that there is a fine grain scale of upgrading still to be explored by
communities and professionals involved with social development, planning and urban design.
Streets are a very important multiple use space that needs to be investigated to unveil
economic, social and spatial potentials to be used to the advantage of inhabitants and of the city
as a whole.

If it is recognized that poverty has a spatial dimension, upgrading will turn to be devised (and
designed) to help the poor to sort out their housing problem and to enhance their life chances
(instead of been historically taken for granted), through more even (and fair) social and spatial
consolidation processes; and might become one more tool to pursue Habitat Agenda
commitments and Millennium Goals. In this sense, the method presented here hopes to
contribute to a review of pre-established informal settlements upgrading standards and indicate
that there are tools available from urban morphology, planning and architecture fields to help
multidisciplinary teams to design tailored interventions for these areas compromised with
inhabitants provision needs, rights of access, values and strategies.

References:

Acioly, C. & Davidson, F. (1998) Densidade Urbana Um instrumento de planejamento e gesto urbana.
Rio de Janeiro: Mau.
Arida, A. (1998) Quantum Environments: Urban Design in the Post-Cartesian Paradigm. Urban Design
International, 3 (3), pp. 141 148.
Bhatt, V. & Scriver, P. (1999) After the Masters, Contemporary Indian Architecture. Ahmedabad: Mapin
Publishing, pp 89 103.
Bhatt, V. (1990) How the Other Half Builds. Montreal: McGill University. V.3.
Brand, S. (1994) How Buildings Learn. London: Union Books, p. 2 22.
Burgess, R. (1985) Problems in the Classification of Low-income Neighbourhoods in Latin America. Third
World Planning Review, v. 7, n. 4, pp 287-306.

23
Cannigia G. & Maffei, L. (1995) Tipologia de la Edificacion, Estructura del Espacio Antropico. Madrid:
Celeste Ediciones S. A.
Cardoso, A (2002a) The Alternative Space: Informal Settlements and Life Chances in Belm, Brazil.
Unpublished PhD Thesis. Oxford: Oxford Brookes University.
Cardoso, A. (2002b) Life Chances and Modern Poverty. Urban Design International, , 223-236.
Castells, M. (2000) Urban Sustainability in the Informational Age. City, Vol. 4, No. 1, pp. 118 122.
Choguill, C. & Choguill, M. (1996) Towards Sustainable Infrastructure for Low-Income Communities. In:
Pugh, C. (ed.) Sustainability, the Environment and Urbanization. London: Earthscan, pp 83 101.
Choguill, C. (1995) The Future of Planned Urban Development in Third World: New Directions. In:
Aldrich, B. & Sandhu, R. (eds.) Housing the Urban Poor. London: Zed Books, 403-414.
Conzen, M. R. G. (1969) Alnwick, Northumberland. A Study in Town Plan Analysis. Oxford, Institute of
British Geographers.
Correa, C. (1985) The New Landscape. Bombay: Book Society of India.
Curtis, W. (1998) Balkrishna Doshi: An Architecture for India. New York: Rizzoli International.
Dahrendorf, R. (1979) Life Chances. London: Weidenfeld and Nicolson.
Dahrendorf, R. (1988) The Modern Social Conflict: An Essay on the Politics of Liberty. London:
Weidenfeld and Nicolson.
Drakakis- Smith, D. (2000) The Third World City. London: Routledge, second edition.
Ford, L. (2000) The Spaces Between Buildings. Baltimore: John Hopkins University Press.
Gilbert, A. & Guggler, J. (2000). Cities, Poverty and Development. Oxford: Oxford University Press,
second edition.
Hanson, J. (2000) Urban Transformations: a History of Design Ideas. Urban Design International 5, 97
122.
Hardoy, J (1989). Squatter Citizen : Life in the Urban Third World. London: Earthscan.
Hardoy, J. & Satterthwaite, D. (1987) The Legal and Illegal City. In: Rodwin, L.(ed) Rodwin, L. (ed.)
Shelter, Settlements and Development, United Nations. Boston/London: Allen & Unwin, pp 304 338.
Hillier, B (1989) The Architecture of the Urban Object. Ekistics. 56, pp. 5-22.
Hillier, B. & Hanson, J. (1984) The Social Logic of Space. Cambridge: Cambridge University Press.
Hillier, B. & Hanson, J. (1998) Space Syntax as a Research Programme. Urban Morphology 2 (2), 108
110.
Hillier, B. (1996) Space is the Machine. Cambridge, Cambridge University Press.
Hillier, B., Greene, M. & Desyllas, J. (2000) Self-generated Neighbourhoods: the Role of Urban Form in
the Consolidation of Informal Settlements. Urban Design International, 5, 61-96.
Holanda, F. (2002) O Espao de Exceo. Brasilia: EdUnB.
Kempen, E. (1994) The Dual City and the Poor: Social Polarisation, Social Segregation and Life Chances.
Urban Studies (31), 995-1015.
Knox, P. & Pinch, S. (2000) Urban Social Geography, an Introduction. Essex: Prentice Hall.
Lemos, C. (1979). Arquitetura Brasileira. So Paulo: Melhoramentos/ Ed. da Universidade de So Paulo.
Levitas, G. (1991) Anthropology and Sociology of Streets. In: Standford, A. (ed) On Streets. Cambridge,
Massachusetts and London: MIT Press, pp 150 - 186.
Lima, J. (2000) Regulatory Instruments and UrbanForm: Searching for Social Equity in Belm, Brazil.
Unpublished PhD thesis. Oxford: Oxford Brookes University.
Lynch, K. (1960) The Image of the City. Cambridge, Massachusetts, and London: MIT Press.
Madaleno, I. (2000) Urban Agriculture in Belm, Brazil. Cities, 17, 1, pp 73 77.
Marx, M. (1991) Cidade no Brasil Terra de Quem? So Paulo: Nobel.

24
McGlynn, 2001, tutorial. Oxford: Oxford Brookes University.
McGlynn, 2002, tutorial. Oxford: Oxford Brookes University.
McGranahan, G. & Satterthwaite, D. (2000) Environmental Health or Ecological Sustainability?
Reconciling the Brown and Green Agendas in Urban Development. In: Pugh, C. (ed.) Sustainable
Cities in Developing Countries. London: Earthscan, pp. 73 90.
Mendes, F. (2003) Avaliao da Ps-Implantao do Sistema de Coleta Condominial, tipo fundo de lote,
do Setor Guanabara, RMB. Master Dissertation Thesis. Belm, PPGEC/UFPA.
Moser, C. (1998) The Asset Vulnerability Framework: Reassessing Urban Poverty Reduction Strategies.
World Development, Vol. 26, No. 1, pp 1-19.
Moudon, A. V. (1997) Urban Morphology as an Emerging Interdisciplinary Field. Urban Morphology, 1, pp
3 10.
Payne, G. (1977) Urban Housing in the Third World. London: Leonard Hill/ Boston: Routledge & Kegan
Paul.
Payne, G. (1984) Low-income Housing in the Developing World. London: John Wiley & Sons.
Payne, G. (1997) Urban Land Tenure and Property Rights in Developing Countries. London: ITP/ODA.
Rapoport, A. (1969) House Form and Culture. Englewood Cliffs, N.J.: Prentice Hall.
Salheen, M. & Forsyth, L. (2001) Addressing Distance in the Space Syntax Syntactical Model. Urban
Design International, 6, 93-110.
Santos, C. N. (1980) Velhas Novidades nos Modos de Urbanizao Brasileiros. In: Valladares, Lcia do
Prado (ed.) Habitao em Questo. Rio de Janeiro: Zahar.
Santos, C. N. (1988) A Cidade como um Jogo de Cartas. Niteri: Universidade Federal Fluminense:
EDUFF; So Paulo: Projeto.
Schiller, S. & Evans, J. (2000) Urban Climate and Compact Cities in Developing Countries. In: Jenks, M &
Burgess, R. (eds.) Compact Cities: Sustainable Urban Forms for Developing Countries. London: Spon,
pp. 117 126.
Sennet, R. (1977) The Fall of Public Man. London/Boston: Faber & Faber.
Soja, E. (2000) Postmetropolis: Critical Studies of Cities and Regions. Oxford: Blackwells.
Steele, J. (1998) The Complete Architecture of Balkrishna Doshi. Rethinking Modernism for the
Developing Country. London: Thames & Hudson, pp. 114 129.
Suttles, G. (1968) The Social Order of the Slum. Chicago: University Press, pp 6-8; 73-78.
Vieira, A. (1997) Housing for Diversity: Roma Gypsies in Madrid. In Beall, J (ed.) A City for All. London,
Zed Books, pp 120 125.
Weber (1964) Order in Diversity, Community without Propinquity. In: L. Wingo, Jr. (ed) Cities and Space:
The Future of Urban Land .Baltimore: John Hopkins University Press.
Williams, K.; Burton, E. & Jenks, M. (2000) Achieving Sustainable Urban Form: Conclusions. In: Williams,
K.; Burton, E. & Jenks, M. (eds.) Sustainable Urban Form. London, E & FN Spon, pp. 347 355.
Wratten, E. (1995) Conceptualizing Urban Poverty. Environment and Urbanization, Vol. 7, No. 1, pp 11
35.
Zeisel, J. (1981) Inquiry by Design: Tools for Environment-Behaviour Research. Belmont: Cambridge
University Press.

25